IFR repair? SPECTRUM ANALYZER ! king has died

[PA0PBZ]

This is how they are connected in the A7750, and they are not even on the +5 rail as you can see.

[SArepairman]

Damnit every lead I tried is bad so far. I took apart the YIG filter can now, and tested some transistors, I got excited because I noticed one of the 4 wire RF transistors was shorted. I desoldered the boards and I found that they grounded 3/4 legs of a transistor. 

Now I need to clean with acetone resolder and reassemble and look for trouble elsewhere.

Do you think that replacing all the op-amp on this board would be problematic as a part of operation clean sweep?
I wanted to replace every single IC component which I can get on digikey cheap and is not programmed.

[SArepairman]

I checked a whole shit load of transistors, of the 66382 type.

Everyone I checked has a Vbe of around 800mV +- 4mV. There is one, in the YIG sweeper, that has a Vbe of 750mV, which is very off compared to all the other ones (I checked like 30 in the log amp, and all the other ones on the YIG assy are all almost 800mV.

Is this type of drift something to suspect? I would have to order this part from those over priced ware house places.

I will desolder and measure to make sure the circuit is not causing a error in the measurement then I will swap it with something that looks unimporant in the log amp lol
can anyone think of a transistor that won't be missed in the log amp schematic?


FUCK THAT GOD DAMN TRANSISTOR WORKS FINE GOD DAMNIT NO NO NO NO 

[SArepairman]

all my dedicated followers  (ok, PA0PBZ gets credit) should be pleased to hear that instead of going on witchhunts I followed the signal and I found a sweep generator that is not working.
There is a overheating HEF4040BP (42C+ with no outputs changing) connected to a circuit with ADC's and DAC.s that are stuck dead in the water.

Considering that many of these parts are legacy and would require ordering from odd vendors at exorbant prices (10 bit ADC for 45$!!!), ad573jn, I began to desolder the parts in the non working circuit block and test them on breadboard.

Hopefully the damage is confined to the HEF4040BP ripple counter, but I need to check near by circuits for operation.
There is also a MK4501N that may be causing trouble.

Thankfully the Mp7531 DAC and the expensive ADC work fine.

Parallel devices are a pleasure to test as you can do it with a simple pulse generator and oscilloscope with no need for any kind of SPI communication or micro controllers.
Maybe some of you have those fancy serial decoders..... hooray old equipment

[PA0PBZ]

Good to see that you are finally getting somewhere. I already thought about telling you that instead of poking around randomly in the poor thing maybe it was time to attack it in a more ehm... structural way, like following the imaginary block diagram and see which blocks are suspicious. Anyway, let's hope that you are on to something and it is not just one of many problems. Time will tell...

[SArepairman]

Well I found one problem with my random approach but it was getting old.


Also I previously wrote:

I also noticed that the board has a screw terminal connector that touches the ground plane by about half a milimeter. When the board is removed the screw terminal does not make a connection to ground, but when the board is inserted and the tension from the housing causes it to buckle a bit the screw terminal grounds that section. I don't know if the engineers wanted to route the ground through the housing at that point... because the ground plane was 95% removed from under the screw terminals, just the corners overlap.... but only make electrical contact while the board is under stress.

Any ideas on this?



the fault appears to have been present since day one.
It seems that it bypasses a star ground. The continuity test shows a connection but of higher resistance .

I just don't understand why the engineers decided to build it so that it only contacts the ground plane when its in the housing with stress applied.

The picture is bad but the corner towards the middle of the board makes contact, not the outer corner as it would seem from the "under tension" part of the diagram. The ground plane was removed from under the suspicious screw terminal in a non square shape, which to me indicates that the layout guy specifically wanted it not to make contact.

The housing for this unit is very robust, I don't think it was damaged.

[PA0PBZ]

The picture is bad

You have no idea... 

[SArepairman]

I can't figure out how the attenuator works in this unit.

The 7550 model uses a relay system for the attenuator, I cracked this can open there are no relays, only some transistors. (I Am waiting for parts and curious).
The 7550 model uses voltage rails to activate these relays, while this model has no inputs (it combined a few of the RF elements which used to be in seperate cans into one big one)....
WTF is going on? Is the YIG assy somehow handling attenuation? The 7550 schematics shows relays which switch in resistor (and RF crap). This thing has no inputs to the attenuator at all. not one wire.

[PA0PBZ]

This thing has no inputs to the attenuator at all. not one wire.

What makes you so sure it is the attenuator?

[SArepairman]

This thing has no inputs to the attenuator at all. not one wire.

What makes you so sure it is the attenuator?

Because it is the first thing that the input goes to.

It looks like some of those PCB capactors/inductors, three 4 legged transistors and some SMD passives.
I fixed my camera so I will take some pictures soon.

Unless the block diagram for the 8000A is different and the attenuator is not the first thing the RF input hard line connects to.

[KJDS]

That does sound like a curious arrangement. I'll wait until I see the pictures before attempting to guess what's there.

[SArepairman]

ah curious, I noticed when I put the center frequency high (~1ghz) the synth lock #1 comes up, only if the bandwidth is narrow, I need to reread the theory of operation I think thats a symptom of something i read..

[PA0PBZ]

This thing has no inputs to the attenuator at all. not one wire.

It looks like some of those PCB capactors/inductors, three 4 legged transistors and some SMD passives.

So if there are no wires, how are you going to power the transistors? My guess is it is some kind of overload protection device.

[KJDS]

This thing has no inputs to the attenuator at all. not one wire.

It looks like some of those PCB capactors/inductors, three 4 legged transistors and some SMD passives.

So if there are no wires, how are you going to power the transistors? My guess is it is some kind of overload protection device.

I didn't read that description too carefully before, but it could well be a limiter. The alternative is that it is powered from a supply on the RF line.

[SArepairman]

This thing has no inputs to the attenuator at all. not one wire.

It looks like some of those PCB capactors/inductors, three 4 legged transistors and some SMD passives.

So if there are no wires, how are you going to power the transistors? My guess is it is some kind of overload protection device.

I didn't read that description too carefully before, but it could well be a limiter. The alternative is that it is powered from a supply on the RF line.

I put the oscilloscope to the power supply rails and I did not see any kind of signals being transmitted when I changed the attenuation settings.

Here is some RF photography:
http://s9.postimg.org/lde3d20jx/DSC_0104.jpg
here is the unit that makes various frequencies and some other stuff

http://s15.postimg.org/3s46ehell/DSC_0106.jpg
the attenuator is on the back side of where the sharpie X is (I need to unscrew it and take pictures of the actual attenuator section but it has the same little transistors you see in the top left quadrant)

here is the interface board:
http://s12.postimg.org/ql5axlun0/DSC_0108.jpg

[KJDS]

http://s15.postimg.org/3s46ehell/DSC_0106.jpg
the attenuator is on the back side of where the sharpie X is (I need to unscrew it and take pictures of the actual attenuator section but it has the same little transistors you see in the top left quadrant)

Those are almost certainly FETs used to amplifier the RF signal

The printed filter in the top left section is a low pass with an added bandstop.

It would really help to have a photo of the other side and it should then be possible to put together a good guess at the block diagram.

[SArepairman]

http://s14.postimg.org/ta2e0wej4/DSC_0113haha.jpg

the hard line input is from the left.

My description was a bit off I was thinking of the previous picture but its pretty similar.
the orange wire is 5.15 volts and it does nothing (signal wise!!).

here is the input attenuator from the 7550, this one makes sense to me.
http://s8.postimg.org/9i61x4bnn/Untitled.png



and the top coaxial connector is for a tracking generator or something

[PA0PBZ]

I'd say this is the first mixer: input LPF, mixer, LO buffer amp (on top, the input is the LO input), BPF and output.
So either someone has stolen your input attenuator, it never had one or you have to find the source of the clicking sound when you adjust the attenuator...

[SArepairman]

I'd say this is the first mixer: input LPF, mixer, LO buffer amp (on top, the input is the LO input), BPF and output.
So either someone has stolen your input attenuator, it never had one or you have to find the source of the clicking sound when you adjust the attenuator...

So the scope has no input attenuator but it has an attenuation function?
wtf??

the scope has no real room for an attenuator, no screw holes for one. I thought it was there to protect the rest of the circuitry?

There is no click when I attenuate.

[PA0PBZ]

That is only 1 of the 3 options 

[SArepairman]

That is only 1 of the 3 options 

?  can attenuation be done afterwords somehow? But it does not click.

[PA0PBZ]

That is only 1 of the 3 options 

?

I said: "either someone has stolen your input attenuator, it never had one or you have to find the source of the clicking sound when you adjust the attenuator..."

Ah, stop editing your posts after I reply please

I think the input attenuator should be the first thing, but who knows what they were thinking. Is there nothing in the users manual about this?

[SArepairman]

That is only 1 of the 3 options 

?

I said: "either someone has stolen your input attenuator, it never had one or you have to find the source of the clicking sound when you adjust the attenuator..."

but it has a attenuation button, which responds in software, and it has no screw holes or anything where a stolen attenuator would go

the manual even says its not an additional feature (the a8000 user manual!!)

[PA0PBZ]

What are the attenuator settings according to the manual? Is it something like 0-60dB in 10 dB steps?

[SArepairman]

What are the attenuator settings according to the manual? Is it something like 0-60dB in 10 dB steps?

yes its 0-60db in 10db steps from the front panel